Local features are not lonely – Laplacian sparse coding for image classification

By Shenghua Gao, Ivor Wai-Hung Tsang, Liang-Tien Chia, Peilin Zhao. 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2010

472 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Media Technology

View Paper: https://www.semanticscholar.org/paper/99b6f85501abfda2b28fde5f05df3792cd91f923

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Media Technology

Citations in the corpus

Robust sparse coding for face recognition (2011, background): “[4] introduced a Laplacian term of coefficient in sparse coding; Wang et al.”
Ask the locals: Multi-way local pooling for image recognition (2011, unknown): “While our method outperforms the Laplacian sparse coding approach [11] on the Caltech 256 dataset, our performance is much below that of Laplacian sparse coding on the Scenes database.”
Robust object tracking via sparsity-based collaborative model (2012, methodology): “Motivated by the success of sparse coding for image classification [30, 24, 9] as well as object tracking [15], we present a generative model for object representation that considers the location information of patches and takes occlusion into account.”
Robust Joint Graph Sparse Coding for Unsupervised Spectral Feature Selection (2017, methodology): “1(c) and Table I] simultaneously conducts manifold learning and a sparse regression to improve the performance in both classification tasks [19] and clustering tasks [14].”
Transfer Sparse Coding for Robust Image Representation (2013, methodology): “To improve the quality of sparse representations, researchers have modified the sparse constraint by adding nonnegative constraint [10], graph regularization [6, 24], weighted l2-norm constraint [20], etc.”
Robust Object Tracking via Sparse Collaborative Appearance Model (2014, methodology): “In this work, we present a collaborative observation model that integrates a discriminative classifier based on holistic templates and a generative model using local representations.”
Deep learning for visual understanding (2017, unknown): “[32] proposed a Laplacian Sparse Coding (LSC) approach, in which similar features are not only assigned to optimallyselected cluster centers, but that also guarantees the selected cluster centers to be similar.”
Sparse Modeling for Image and Vision Processing (2014, background): “This is precisely what Boureau et al. [2011] have proposed to do [see also Gao et al., 2010, 2013, Wang et al., 2010, Yang et al., 2012b, Zhou et al., 2010, Koniusz and Mikolajczyk, 2011, for related work that will be partly detailed in the next section].”
Unsupervised Cross-Dataset Transfer Learning for Person Re-identification (2016, background): “Graph Laplacian regularisation has also been explored in a sparse coding formulation before, for problems such as unsupervised clustering [8, 42], or supervised face verification/recognition [13].”
Deep learning for visual understanding: A review (2016, methodology): “Deep Belief Networks[49] Deep Boltzmann Machines[65] Deep Energy Models[73] Sparse Autoencoder[50] Denoising Autoencoder[85] Contractive Autoencoder[87] Sparse Coding SPM[98] Laplacian Sparse Coding[113] Local Coordinate Coding[95] Super-Vector Coding[118] AlexNet[6] Clarifai[52] VGG[31] GoogLeNet[2”
Laplacian Sparse Coding, Hypergraph Laplacian Sparse Coding, and Applications (2013, result): “This paper is an extension of our previous work [15].”
Graph Regularized Sparse Coding for Image Representation (2011, background): “It is important to note that the similar idea has also been proposed in [34].”
In defense of soft-assignment coding (2011, result): “Its classification accuracy is also comparable to that obtained by using the sparse coding, as reported in [5].”
Group Sparsity and Geometry Constrained Dictionary Learning for Action Recognition from Depth Maps (2013, methodology): “[5], we propose to add geometry constraint to the class-specific dictionary learning process.”
Image classification by non-negative sparse coding, low-rank and sparse decomposition (2011, methodology): “The codebook size for non-negative sparse coding is set to 1,024, as in [4, 16, 28].”